Multifunctional Ca2+/calmodulin-dependent protein kinase (CaM kinase) is a major protein kinase orchestrating the physiological effects of hormones, neurotransmitters and growth factors in numerous tissues. It responds to stimuli that elevate intracellular Ca2+ and phosphorylates substrates which are localized in the cytosol, membrane and cytoskeletal compartments. Its designation as a general protein kinase brings us to an important question--how does a cell utilize CaM kinase as a mediator of multiple Ca2+-linked hormones, yet still achieve specificity in its cellular responses? We propose to test the hypothesis that the autoregulatory properties of CaM kinase enable it to decode the frequency of Ca2+ oscillations and spikes, thus adding a novel form of temporal response specificity. Furthermore, we propose to test the hypothesis that spatial targeting of CaM kinase isoforms to distinct intracellular sites helps to determine site-specific effects of hormones. Temporal regulation of this multimeric enzyme occurs by its unique autophosphorylation properties. We will extend our studies demonstrating that autophosphorylation occurs when an activated subunit phosphorylates a proximate neighbor in the holoenzyme which also has calmodulin bound. These cooperative properties will be elaborated, since they form the basis for a model by which the kinase may be selectively activated at appropriate Ca2+ spike frequencies. The model will be tested in vitro using immobilized kinase exposed to rapid fluctuations in Ca2+ and in situ by testing frequency dependent activation of the kinase and of the Cl-channel that it regulates in epithelial cells. One of our newly cloned CaM kinase isoforms is targeted to the nucleus and others to the cytoskeleton suggesting a spatial basis for response specificity. We will identify the position of catalytic/regulatory and targeting domains within the structure by electron microscopy. We will identify the nuclear localization sequence and determine whether the enzyme enters the nucleus as a monomeric intermediate or as a multimer. Targeting of the cytoskeletal isoform and its translocation by activation will also be examined. With this basic knowledge, we will target engineered constructs to the nucleus and to the cytoskeleton and use these as reporters of functional Ca2+/calmodulin elevated at these sites by several distinct Ca2-linked signal transduction pathways.